IV-SAKG: Industrial Control System Vulnerability Knowledge Graph

IV-SAKG is a scenario-aware knowledge graph dataset constructed for vulnerability reasoning in industrial control systems. It contains entities, relations, and triples built upon public vulnerability sources like CVE, CISA, and CNVD. The dataset was authored by ling zhang and last updated on 2026-05-25.

Use Cases

• Knowledge graph completion tasks based on the provided entity-relation triples.
• Vulnerability impact assessment based on the scenario-aware graph structure.
• Comparative analysis of vulnerability representations using the provided full and base dataset versions.
• Training and evaluating graph-based machine learning models using the provided train/valid/test splits.

Strengths

• Dataset size is 26.1 MB, providing a specific scale for download planning.
• Two distinct versions (IV-SAKG-full and IV-SAKG-base) are provided to facilitate ablation studies.
• Data is pre-split into training, validation, and test subsets for machine learning tasks.
• Data is sourced from public vulnerability databases (CVE, CISA, CNVD).

Limitations

• Row count is unknown, which may limit suitability assessment for specific modeling needs.
• Column-level documentation is absent; field semantics must be inferred after download.
• Description metadata is limited; actual data quality requires manual inspection after download.

Provenance

Source

Public vulnerability sources (CVE, CISA, CNVD).

Collection Method

Constructed as a knowledge graph from the mentioned sources.

Freshness

Last updated 2026-05-25 08:18:40; freshness should be verified.